DonorsChoose¶
DonorsChoose.org receives hundreds of thousands of project proposals each year for classroom projects in need of funding. Right now, a large number of volunteers is needed to manually screen each submission before it's approved to be posted on the DonorsChoose.org website.
Next year, DonorsChoose.org expects to receive close to 500,000 project proposals. As a result, there are three main problems they need to solve:
- How to scale current manual processes and resources to screen 500,000 projects so that they can be posted as quickly and as efficiently as possible
- How to increase the consistency of project vetting across different volunteers to improve the experience for teachers
- How to focus volunteer time on the applications that need the most assistance
The goal of the competition is to predict whether or not a DonorsChoose.org project proposal submitted by a teacher will be approved, using the text of project descriptions as well as additional metadata about the project, teacher, and school. DonorsChoose.org can then use this information to identify projects most likely to need further review before approval.
About the DonorsChoose Data Set¶
The train.csv data set provided by DonorsChoose contains the following features:
| Feature | Description |
|---|---|
project_id |
A unique identifier for the proposed project. Example: p036502 |
project_title |
Title of the project. Examples:
|
project_grade_category |
Grade level of students for which the project is targeted. One of the following enumerated values:
|
project_subject_categories |
One or more (comma-separated) subject categories for the project from the following enumerated list of values:
Examples:
|
school_state |
State where school is located (Two-letter U.S. postal code). Example: WY |
project_subject_subcategories |
One or more (comma-separated) subject subcategories for the project. Examples:
|
project_resource_summary |
An explanation of the resources needed for the project. Example:
|
project_essay_1 |
First application essay* |
project_essay_2 |
Second application essay* |
project_essay_3 |
Third application essay* |
project_essay_4 |
Fourth application essay* |
project_submitted_datetime |
Datetime when project application was submitted. Example: 2016-04-28 12:43:56.245 |
teacher_id |
A unique identifier for the teacher of the proposed project. Example: bdf8baa8fedef6bfeec7ae4ff1c15c56 |
teacher_prefix |
Teacher's title. One of the following enumerated values:
|
teacher_number_of_previously_posted_projects |
Number of project applications previously submitted by the same teacher. Example: 2 |
* See the section Notes on the Essay Data for more details about these features.
Additionally, the resources.csv data set provides more data about the resources required for each project. Each line in this file represents a resource required by a project:
| Feature | Description |
|---|---|
id |
A project_id value from the train.csv file. Example: p036502 |
description |
Desciption of the resource. Example: Tenor Saxophone Reeds, Box of 25 |
quantity |
Quantity of the resource required. Example: 3 |
price |
Price of the resource required. Example: 9.95 |
Note: Many projects require multiple resources. The id value corresponds to a project_id in train.csv, so you use it as a key to retrieve all resources needed for a project:
The data set contains the following label (the value you will attempt to predict):
| Label | Description |
|---|---|
project_is_approved |
A binary flag indicating whether DonorsChoose approved the project. A value of 0 indicates the project was not approved, and a value of 1 indicates the project was approved. |
Notes on the Essay Data¶
-
Prior to May 17, 2016, the prompts for the essays were as follows:
- __project_essay_1:__ "Introduce us to your classroom"
- __project_essay_2:__ "Tell us more about your students"
- __project_essay_3:__ "Describe how your students will use the materials you're requesting"
- __project_essay_3:__ "Close by sharing why your project will make a difference"
-
Starting on May 17, 2016, the number of essays was reduced from 4 to 2, and the prompts for the first 2 essays were changed to the following:
- __project_essay_1:__ "Describe your students: What makes your students special? Specific details about their background, your neighborhood, and your school are all helpful."
- __project_essay_2:__ "About your project: How will these materials make a difference in your students' learning and improve their school lives?"
For all projects with project_submitted_datetime of 2016-05-17 and later, the values of project_essay_3 and project_essay_4 will be NaN.
In [2]:
import os
os.chdir("F:/f/MY____/AAIC/AssignmentS/DonorsChoose_Data")
os.getcwd()
Out[2]:
In [3]:
%matplotlib inline
import warnings
warnings.filterwarnings("ignore")
import sqlite3
import pandas as pd
import numpy as np
import nltk
import string
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics import confusion_matrix
from sklearn import metrics
from sklearn.metrics import roc_curve, auc
from nltk.stem.porter import PorterStemmer
import re
# Tutorial about Python regular expressions: https://pymotw.com/2/re/
import string
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer
from nltk.stem.wordnet import WordNetLemmatizer
from gensim.models import Word2Vec
from gensim.models import KeyedVectors
import pickle
from tqdm import tqdm
import os
from plotly import plotly
import plotly.offline as offline
import plotly.graph_objs as go
offline.init_notebook_mode()
from collections import Counter
1.1 Reading Data¶
In [4]:
project_data = pd.read_csv('train_data.csv')
resource_data = pd.read_csv('resources.csv')
In [5]:
print("Number of data points in train data", project_data.shape)
print('-'*50)
print("The attributes of data :", project_data.columns.values)
In [6]:
print("Number of data points in train data", resource_data.shape)
print(resource_data.columns.values)
resource_data.head(2)
Out[6]:
1.2 preprocessing of project_subject_categories¶
In [7]:
catogories = list(project_data['project_subject_categories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_') # we are replacing the & value into
cat_list.append(temp.strip())
project_data['clean_categories'] = cat_list
project_data.drop(['project_subject_categories'], axis=1, inplace=True)
from collections import Counter
my_counter = Counter()
for word in project_data['clean_categories'].values:
my_counter.update(word.split())
cat_dict = dict(my_counter)
sorted_cat_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))
1.3 preprocessing of project_subject_subcategories¶
In [8]:
sub_catogories = list(project_data['project_subject_subcategories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
sub_cat_list = []
for i in sub_catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp +=j.strip()+" "#" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_')
sub_cat_list.append(temp.strip())
project_data['clean_subcategories'] = sub_cat_list
project_data.drop(['project_subject_subcategories'], axis=1, inplace=True)
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
my_counter = Counter()
for word in project_data['clean_subcategories'].values:
my_counter.update(word.split())
sub_cat_dict = dict(my_counter)
sorted_sub_cat_dict = dict(sorted(sub_cat_dict.items(), key=lambda kv: kv[1]))
1.3 Text preprocessing¶
In [23]:
# merge two column text dataframe:
project_data["essay"] = project_data["project_essay_1"].map(str) +\
project_data["project_essay_2"].map(str) + \
project_data["project_essay_3"].map(str) + \
project_data["project_essay_4"].map(str)
In [10]:
project_data.head(2)
Out[10]:
In [11]:
# printing some random reviews
print(project_data['essay'].values[0])
print("="*50)
print(project_data['essay'].values[150])
print("="*50)
print(project_data['essay'].values[1000])
print("="*50)
print(project_data['essay'].values[20000])
print("="*50)
print(project_data['essay'].values[99999])
print("="*50)
Introducing New Features¶
[Task-2] Apply Logistic Regression on the below feature set Set 5 by finding the best hyper parameter GridSearch </br>
Consider these set of features for Set 5 in Assignment:
categorical dataschool_state clean_categories....clean_subcategories....project_grade_category....teacher_prefix
numerical data quantity....teacher_number_of_previously_posted_projects....price
New Features
sentiment score's of each of the essay : numerical data
number of words in the title : numerical data
number of words in the combine essays : numerical data
In [20]:
new_title = []
for i in tqdm(project_data['project_title']):
j = decontracted(i)
new_title.append(j)
In [21]:
#Introducing New Features
title_word_count = []
#for i in project_data['project_title']:
for i in tqdm(new_title):
j = len(i.split())
title_word_count.append(j)
#print(j)
project_data['title_word_count'] = title_word_count
In [22]:
project_data.head(2)
Out[22]:
In [24]:
new_essay = []
for i in tqdm(project_data['essay']):
j = decontracted(i)
new_essay.append(j)
In [26]:
essay_word_count = []
for i in tqdm(new_essay):
j = len(i.split())
essay_word_count.append(j)
#print(j)
project_data['essay_word_count'] = essay_word_count
In [27]:
project_data.head(2)
Out[27]:
Computing Sentiment Scores¶
In [28]:
import nltk
from nltk.sentiment.vader import SentimentIntensityAnalyzer
# import nltk
#nltk.download('vader_lexicon')
sid = SentimentIntensityAnalyzer()
for_sentiment = 'a person is a person no matter how small dr seuss i teach the smallest students with the biggest enthusiasm'
ss = sid.polarity_scores(for_sentiment)
for k in ss:
print('{0}: {1}, '.format(k, ss[k]), end='')
# we can use these 4 things as features/attributes (neg, neu, pos, compound)
# neg: 0.0, neu: 0.753, pos: 0.247, compound: 0.93
In [29]:
SID = SentimentIntensityAnalyzer()
#There is NEGITIVE and POSITIVE and NEUTRAL and COMPUND SCORES
#http://www.nltk.org/howto/sentiment.html
negitive = []
positive = []
neutral = []
compound = []
for i in tqdm(project_data['essay']):
j = SID.polarity_scores(i)['neg']
k = SID.polarity_scores(i)['neu']
l = SID.polarity_scores(i)['pos']
m = SID.polarity_scores(i)['compound']
negitive.append(j)
positive.append(k)
neutral.append(l)
compound.append(m)
In [ ]:
project_data['negitive'] = negitive
In [32]:
project_data['positive'] = positive
project_data['neutral'] = neutral
project_data['compound'] = compound
project_data.head(2)
Out[32]:
1.4 Train_Test_Split¶
In [33]:
#Train Test Split
from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(project_data, project_data["project_is_approved"],
test_size = 0.33, stratify = project_data["project_is_approved"],
random_state = 42)
In [34]:
#Train CV Split
x_train, x_cv, y_train, y_cv = train_test_split(x_train, y_train, test_size = 0.33, stratify = y_train,
random_state = 42)
In [35]:
print(x_test.columns)
print(x_train.columns)
#print(x_cv.columns)
#print(x_train.shape)
#print(x_test.shape)
#print(x_cv.shape)
In [36]:
#Dropping Class Label in train test and cv data
x_train.drop(["project_is_approved"], axis = 1, inplace = True)
x_test.drop(["project_is_approved"], axis = 1, inplace = True)
x_cv.drop(["project_is_approved"], axis = 1, inplace = True)
In [37]:
print(x_train.columns)
In [ ]:
Preparing Data for Models¶
In [38]:
# https://stackoverflow.com/a/47091490/4084039
import re
def decontracted(phrase):
# specific
phrase = re.sub(r"won't", "will not", phrase)
phrase = re.sub(r"can\'t", "can not", phrase)
# general
phrase = re.sub(r"n\'t", " not", phrase)
phrase = re.sub(r"\'re", " are", phrase)
phrase = re.sub(r"\'s", " is", phrase)
phrase = re.sub(r"\'d", " would", phrase)
phrase = re.sub(r"\'ll", " will", phrase)
phrase = re.sub(r"\'t", " not", phrase)
phrase = re.sub(r"\'ve", " have", phrase)
phrase = re.sub(r"\'m", " am", phrase)
return phrase
In [39]:
sent = decontracted(project_data['essay'].values[20000])
print(sent)
print("="*50)
In [40]:
# \r \n \t remove from string python: http://texthandler.com/info/remove-line-breaks-python/
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
print(sent)
In [41]:
#remove spacial character: https://stackoverflow.com/a/5843547/4084039
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
print(sent)
In [42]:
# https://gist.github.com/sebleier/554280
# we are removing the words from the stop words list: 'no', 'nor', 'not'
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
"you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their',\
'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those', \
'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \
'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \
'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\
'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\
'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\
'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm', 'o', 're', \
've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't", 'hadn',\
"hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn',\
"mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren', "weren't", \
'won', "won't", 'wouldn', "wouldn't"]
Project_Essays preprocessing¶
In [43]:
#train_preprocessed_essays
# Combining all the above stundents
from tqdm import tqdm
train_preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(x_train['essay'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
train_preprocessed_essays.append(sent.lower().strip())
In [44]:
train_preprocessed_essays[10]
Out[44]:
In [45]:
# test_preprocessed_essay
from tqdm import tqdm
test_preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(x_test['essay'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
test_preprocessed_essays.append(sent.lower().strip())
In [47]:
test_preprocessed_essays[10]
Out[47]:
In [48]:
# CV_preprocessed_essays
from tqdm import tqdm
cv_preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(x_cv['essay'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
cv_preprocessed_essays.append(sent.lower().strip())
In [49]:
cv_preprocessed_essays[10]
Out[49]:
Project_Titles_preprocessing¶
In [50]:
# train_preprocessed_title
from tqdm import tqdm
train_preprocessed_titles = []
# tqdm is for printing the status bar
for sentance in tqdm(x_train['project_title'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
train_preprocessed_titles.append(sent.lower().strip())
In [51]:
train_preprocessed_titles[10]
Out[51]:
In [52]:
# Test_preprocessed_essays
from tqdm import tqdm
test_preprocessed_titles = []
# tqdm is for printing the status bar
for sentance in tqdm(x_test['project_title'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
test_preprocessed_titles.append(sent.lower().strip())
In [53]:
test_preprocessed_titles[10]
Out[53]:
In [54]:
# CV_preprocessed_titles
from tqdm import tqdm
cv_preprocessed_titles = []
# tqdm is for printing the status bar
for sentance in tqdm(x_cv['project_title'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
cv_preprocessed_titles.append(sent.lower().strip())
In [55]:
# after preprocesing
cv_preprocessed_titles[10]
Out[55]:
In [56]:
project_data.columns
Out[56]:
we are going to consider
- school_state : categorical data
- clean_categories : categorical data
- clean_subcategories : categorical data
- project_grade_category : categorical data
- teacher_prefix : categorical data
- project_title : text data
- text : text data
- project_resource_summary: text data (optinal)
- quantity : numerical (optinal)
- teacher_number_of_previously_posted_projects : numerical
- price : numerical1.5.1 Vectorizing Categorical data¶
In [57]:
# we use count vectorizer to convert the values into one
# Vectorizing Clean Categories
from sklearn.feature_extraction.text import CountVectorizer
vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(x_train['clean_categories'].values)
train_categories_one_hot = vectorizer.transform(x_train['clean_categories'].values)
test_categories_one_hot = vectorizer.transform(x_test['clean_categories'].values)
cv_categories_one_hot = vectorizer.transform(x_cv['clean_categories'].values)
print(vectorizer.get_feature_names())
print("Shape of Train matrix after one hot encodig ",train_categories_one_hot.shape)
print("Shape of Test matrix after one hot encodig ",test_categories_one_hot.shape)
print("Shape of cv matrix after one hot encodig ",cv_categories_one_hot.shape)
In [58]:
# we use count vectorizer to convert the values into one
vectorizer = CountVectorizer(vocabulary=list(sorted_sub_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(x_train["clean_subcategories"].values)
train_sub_categories_one_hot = vectorizer.transform(x_train['clean_subcategories'].values)
test_sub_categories_one_hot = vectorizer.transform(x_test['clean_subcategories'].values)
cv_sub_categories_one_hot = vectorizer.transform(x_cv['clean_subcategories'].values)
print(vectorizer.get_feature_names())
print("Shape of Train matrix after Trainone hot encodig ",train_sub_categories_one_hot.shape)
print("Shape of test matrix after one hot encodig ",test_sub_categories_one_hot.shape)
print("Shape of cv_ matrix after one hot encodig ",cv_sub_categories_one_hot.shape)
In [59]:
# you can do the similar thing with state, teacher_prefix and project_grade_category also
from collections import Counter
my_counter = Counter()
for word in project_data["school_state"].values:
my_counter.update(word.split())
In [60]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
state_cat_dict = dict(my_counter)
storted_state_cat_dict = dict(sorted(state_cat_dict.items(), key=lambda kv: kv[1]))
In [61]:
# Please do the similar feature encoding with state, teacher_prefix and project_grade_category also
#Using Count Vectorizer to convert the state value onto on hot encoded feature
vectorizer = CountVectorizer(vocabulary=list(storted_state_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['school_state'].values)
print(vectorizer.get_feature_names())
train_state_one_hot = vectorizer.transform(x_train['school_state'].values)
test_state_one_hot = vectorizer.transform(x_test['school_state'].values)
cv_state_one_hot = vectorizer.transform(x_cv['school_state'].values)
print("Shape of Train matrix after one hot encodig ",train_state_one_hot.shape)
print("Shape of Test matrix after one hot encodig ",test_state_one_hot.shape)
print("Shape of CV matrix after one hot encodig ",cv_state_one_hot.shape)
In [62]:
#https://stackoverflow.com/questions/42224700/attributeerror-float-object-has-no-attribute-split
project_data['project_grade_category']=project_data['project_grade_category'].fillna("")
my_counter = Counter()
for word in project_data['project_grade_category'].values:
my_counter.update(word.split())
In [63]:
project_cat_dict = dict(my_counter)
sorted_project_cat_dict = dict(sorted(project_cat_dict.items(), key=lambda kv: kv[1]))
In [64]:
# feature encoding for project_grade_category also
vectorizer = CountVectorizer(vocabulary=list(sorted_project_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['project_grade_category'].values)
print(vectorizer.get_feature_names())
train_grade_one_hot = vectorizer.transform(x_train['project_grade_category'].values)
test_grade_one_hot = vectorizer.transform(x_test['project_grade_category'].values)
cv_grade_one_hot = vectorizer.transform(x_cv['project_grade_category'].values)
print("Shape of Train matrix after one hot encodig ",train_grade_one_hot.shape)
print("Shape of test matrix after one hot encodig ",test_grade_one_hot.shape)
print("Shape of cv matrix after one hot encodig ",cv_grade_one_hot.shape)
In [65]:
#https://stackoverflow.com/questions/42224700/attributeerror-float-object-has-no-attribute-split
project_data['teacher_prefix']=project_data['teacher_prefix'].fillna(" ")
In [66]:
my_counter = Counter()
for word in project_data['teacher_prefix'].values:
my_counter.update(word.split())
In [67]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
teacher_cat_dict = dict(my_counter)
sorted_teacher_cat_dict = dict(sorted(teacher_cat_dict.items(), key=lambda kv: kv[1]))
In [68]:
#Using Count Vectorizer to convert the teacher_prefix value onto on hot encoded feature
#ValueError: np.nan is an invalid document, expected byte or unicode string.
#https://stackoverflow.com/questions/39303912/tfidfvectorizer-in-scikit-learn-valueerror-np-nan-is-an-invalid-document
vectorizer = CountVectorizer(vocabulary=list(sorted_teacher_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['teacher_prefix'].values.astype('U'))
print(vectorizer.get_feature_names())
train_teacher_one_hot = vectorizer.transform(x_train['teacher_prefix'].values.astype('U'))
test_teacher_one_hot = vectorizer.transform(x_test['teacher_prefix'].values.astype('U'))
cv_teacher_one_hot = vectorizer.transform(x_cv['teacher_prefix'].values.astype('U'))
print("Shape of Train matrix after one hot encodig ",train_teacher_one_hot.shape)
print("Shape of Test matrix after one hot encodig ",test_teacher_one_hot.shape)
print("Shape of CV matrix after one hot encodig ",cv_teacher_one_hot.shape)
1.5.2 Vectorizing Text data¶
1.5.2.1 Bag of words¶
BI-Grams of minimum frequency of words 10 (min_df = 10) of 5000 features (max_features = 5000)
In [69]:
# We are considering only the words which appeared in at least 10 documents(rows or projects).
vectorizer = CountVectorizer(ngram_range = (2,2), max_features = 5000, min_df=10)
vectorizer.fit(train_preprocessed_essays)
train_text_bow = vectorizer.transform(train_preprocessed_essays)
print("Shape of matrix after one hot encodig ",train_text_bow.shape)
In [70]:
#Vectorizing Test Data
# you can vectorize the title also
# before you vectorize the title make sure you preprocess it
test_text_bow = vectorizer.transform(test_preprocessed_essays)
print("Shape of matrix after one hot encodig ",test_text_bow.shape)
In [71]:
# Vectrozing CV Data
cv_text_bow = vectorizer.transform(cv_preprocessed_essays)
print("Shape of matrix after one hot encodig ",cv_text_bow.shape)
Project_Title BOW¶
In [72]:
# We are considering only the words which appeared in at least 10 documents(rows or projects).
vectorizer = CountVectorizer(ngram_range = (2,2), max_features = 5000, min_df=10)
vectorizer.fit(train_preprocessed_titles)
train_titles_bow = vectorizer.transform(train_preprocessed_titles)
print("Shape of matrix after one hot encodig ",train_titles_bow.shape)
In [73]:
#Vectorizing Test Data
test_titles_bow = vectorizer.transform(test_preprocessed_titles)
print("Shape of matrix after one hot encodig ",test_titles_bow.shape)
In [74]:
#Vectrizing CV Data
cv_titles_bow = vectorizer.transform(cv_preprocessed_titles)
print("Shape of matrix after one hot encodig ",cv_titles_bow.shape)
1.5.2.2 TFIDF vectorizer¶
Here also we are considering the Bi-Grams of word frequency of 10 words(min_df = 10) and Maximum features 5000
In [75]:
from sklearn.feature_extraction.text import TfidfVectorizer
vectorizer = TfidfVectorizer(min_df=10, max_features = 5000, ngram_range = (2,2))
vectorizer.fit(train_preprocessed_essays)
train_text_tfidf = vectorizer.transform(train_preprocessed_essays)
print("Shape of matrix after one hot encodig ",train_text_tfidf.shape)
In [76]:
test_text_tfidf = vectorizer.transform(test_preprocessed_essays)
print("Shape of matrix after one hot encodig ",test_text_tfidf.shape)
In [77]:
cv_text_tfidf = vectorizer.transform(cv_preprocessed_essays)
print("Shape of matrix after one hot encodig ",cv_text_tfidf.shape)
Project Titles¶
In [78]:
vectorizer = TfidfVectorizer(min_df=10, max_features = 5000, ngram_range = (2,2))
vectorizer.fit(train_preprocessed_titles)
train_title_tfidf = vectorizer.transform(train_preprocessed_titles)
print("Shape of matrix after one hot encodig ",train_title_tfidf.shape)
In [79]:
test_title_tfidf = vectorizer.transform(test_preprocessed_titles)
print("Shape of matrix after one hot encodig ",test_title_tfidf.shape)
In [80]:
cv_title_tfidf = vectorizer.transform(cv_preprocessed_titles)
print("Shape of matrix after one hot encodig ",cv_title_tfidf.shape)
1.5.2.3 Using Pretrained Models: Avg W2V¶
In [81]:
'''
# Reading glove vectors in python: https://stackoverflow.com/a/38230349/4084039
def loadGloveModel(gloveFile):
print ("Loading Glove Model")
f = open(gloveFile,'r', encoding="utf8")
model = {}
for line in tqdm(f):
splitLine = line.split()
word = splitLine[0]
embedding = np.array([float(val) for val in splitLine[1:]])
model[word] = embedding
print ("Done.",len(model)," words loaded!")
return model
model = loadGloveModel('glove.42B.300d.txt')
# ============================
Output:
Loading Glove Model
1917495it [06:32, 4879.69it/s]
Done. 1917495 words loaded!
# ============================
words = []
for i in preproced_texts:
words.extend(i.split(' '))
for i in preproced_titles:
words.extend(i.split(' '))
print("all the words in the coupus", len(words))
words = set(words)
print("the unique words in the coupus", len(words))
inter_words = set(model.keys()).intersection(words)
print("The number of words that are present in both glove vectors and our coupus", \
len(inter_words),"(",np.round(len(inter_words)/len(words)*100,3),"%)")
words_courpus = {}
words_glove = set(model.keys())
for i in words:
if i in words_glove:
words_courpus[i] = model[i]
print("word 2 vec length", len(words_courpus))
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
import pickle
with open('glove_vectors', 'wb') as f:
pickle.dump(words_courpus, f)
'''
Out[81]:
In [82]:
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
# make sure you have the glove_vectors file
with open('glove_vectors', 'rb') as f:
model = pickle.load(f)
glove_words = set(model.keys())
In [83]:
# average Word2Vec
# compute average word2vec for each review.
train_avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(train_preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
train_avg_w2v_vectors.append(vector)
print(len(train_avg_w2v_vectors))
print(len(train_avg_w2v_vectors[0]))
In [84]:
# average Word2Vec
# compute average word2vec for each review.
test_avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(test_preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
test_avg_w2v_vectors.append(vector)
print(len(test_avg_w2v_vectors))
print(len(test_avg_w2v_vectors[0]))
In [85]:
# average Word2Vec
# compute average word2vec for each review.
cv_avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(cv_preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
cv_avg_w2v_vectors.append(vector)
print(len(cv_avg_w2v_vectors))
print(len(cv_avg_w2v_vectors[0]))
AVG_W2V Project_Titles¶
In [86]:
# average Word2Vec
# compute average word2vec for each review.
train_title_avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(train_preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
train_title_avg_w2v_vectors.append(vector)
print(len(train_title_avg_w2v_vectors))
print(len(train_title_avg_w2v_vectors[0]))
In [87]:
# average Word2Vec
# compute average word2vec for each review.
test_title_avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(test_preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
test_title_avg_w2v_vectors.append(vector)
print(len(test_title_avg_w2v_vectors))
print(len(test_title_avg_w2v_vectors[0]))
In [88]:
# average Word2Vec
# compute average word2vec for each review.
cv_title_avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(cv_preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
cv_title_avg_w2v_vectors.append(vector)
print(len(cv_title_avg_w2v_vectors))
print(len(cv_title_avg_w2v_vectors[0]))
1.5.2.3 Using Pretrained Models: TFIDF weighted W2V¶
In [89]:
# S = ["abc def pqr", "def def def abc", "pqr pqr def"]
tfidf_model = TfidfVectorizer()
tfidf_model.fit(train_preprocessed_essays)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [90]:
# average Word2Vec
# compute average word2vec for each review.
train_essay_tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(train_preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
train_essay_tfidf_w2v_vectors.append(vector)
print(len(train_essay_tfidf_w2v_vectors))
print(len(train_essay_tfidf_w2v_vectors[0]))
In [91]:
# Similarly you can vectorize for title also
# average Word2Vec
# compute average word2vec for each review.
test_essay_tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(test_preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
test_essay_tfidf_w2v_vectors.append(vector)
print(len(test_essay_tfidf_w2v_vectors))
print(len(test_essay_tfidf_w2v_vectors[0]))
In [92]:
# average Word2Vec
# compute average word2vec for each review.
cv_essay_tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(cv_preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
cv_essay_tfidf_w2v_vectors.append(vector)
print(len(cv_essay_tfidf_w2v_vectors))
print(len(cv_essay_tfidf_w2v_vectors[0]))
Proect_titles¶
In [93]:
# average Word2Vec
# compute average word2vec for each review.
train_title_tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(train_preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
train_title_tfidf_w2v_vectors.append(vector)
print(len(train_title_tfidf_w2v_vectors))
print(len(train_title_tfidf_w2v_vectors[0]))
In [94]:
# average Word2Vec
# compute average word2vec for each review.
test_title_tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(test_preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
test_title_tfidf_w2v_vectors.append(vector)
print(len(test_title_tfidf_w2v_vectors))
print(len(test_title_tfidf_w2v_vectors[0]))
In [95]:
# average Word2Vec
# compute average word2vec for each review.
cv_title_tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(cv_preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
cv_title_tfidf_w2v_vectors.append(vector)
print(len(cv_title_tfidf_w2v_vectors))
print(len(cv_title_tfidf_w2v_vectors[0]))
1.5.3 Vectorizing Numerical features¶
In [96]:
price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
project_data = pd.merge(project_data, price_data, on='id', how='left')
print(price_data.head())
#print(project_data.columns)
print(x_train.columns)
In [97]:
# - quantity : numerical (optinal)
# - teacher_number_of_previously_posted_projects : numerical
# - price : numerical
x_train = pd.merge(x_train, price_data, on = "id", how = "left")
#print(x_train.columns)
x_test = pd.merge(x_test, price_data, on = "id", how = "left")
x_cv = pd.merge(x_cv, price_data, on = "id", how = "left")
Standardize Price¶
In [98]:
# check this one: https://www.youtube.com/watch?v=0HOqOcln3Z4&t=530s
# standardization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html
from sklearn.preprocessing import StandardScaler
# price_standardized = standardScalar.fit(project_data['price'].values)
# this will rise the error
# ValueError: Expected 2D array, got 1D array instead: array=[725.05 213.03 329. ... 399. 287.73 5.5 ].
# Reshape your data either using array.reshape(-1, 1)
price_scalar = StandardScaler()
price_scalar.fit(x_train['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
train_price_standar = price_scalar.transform(x_train['price'].values.reshape(-1, 1))
train_price_standar
Out[98]:
In [99]:
price_scalar.fit(x_test['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
test_price_standar = price_scalar.transform(x_test['price'].values.reshape(-1, 1))
test_price_standar
Out[99]:
In [100]:
price_scalar.fit(x_cv['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
cv_price_standar = price_scalar.transform(x_cv['price'].values.reshape(-1, 1))
test_price_standar
Out[100]:
In [101]:
print(train_price_standar.shape, y_train.shape)
print(test_price_standar.shape, y_test.shape)
print(cv_price_standar.shape, y_cv.shape)
Standardize Teacher previously posted Projects¶
In [102]:
warnings.filterwarnings("ignore")
price_scalar.fit(x_train['teacher_number_of_previously_posted_projects'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
train_prev_proj_standar = price_scalar.transform(x_train['teacher_number_of_previously_posted_projects'].values.reshape(-1, 1))
train_prev_proj_standar
Out[102]:
In [103]:
price_scalar.fit(x_test['teacher_number_of_previously_posted_projects'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
test_prev_proj_standar = price_scalar.transform(x_test['teacher_number_of_previously_posted_projects'].values.reshape(-1, 1))
test_prev_proj_standar
Out[103]:
In [104]:
price_scalar.fit(x_cv['teacher_number_of_previously_posted_projects'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
cv_prev_proj_standar = price_scalar.transform(x_cv['teacher_number_of_previously_posted_projects'].values.reshape(-1, 1))
cv_prev_proj_standar
Out[104]:
In [105]:
print(train_prev_proj_standar.shape, y_train.shape)
print(test_prev_proj_standar.shape, y_test.shape)
print(cv_prev_proj_standar.shape, y_cv.shape)
Standardize Quantity¶
In [106]:
price_scalar.fit(x_train['quantity'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
train_quantity_standar = price_scalar.transform(x_train['quantity'].values.reshape(-1, 1))
train_quantity_standar
Out[106]:
In [107]:
price_scalar.fit(x_test['quantity'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
test_quantity_standar = price_scalar.transform(x_test['quantity'].values.reshape(-1, 1))
test_quantity_standar
Out[107]:
In [108]:
price_scalar.fit(x_cv['quantity'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
cv_quantity_standar = price_scalar.transform(x_cv['quantity'].values.reshape(-1, 1))
cv_quantity_standar
Out[108]:
In [109]:
print(train_quantity_standar.shape, y_train.shape)
print(test_quantity_standar.shape, y_test.shape)
print(cv_quantity_standar.shape, y_cv.shape)
Standardize Title_word_count¶
In [110]:
title_scalar = StandardScaler()
title_scalar.fit(x_train['title_word_count'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {title_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
train_title_word_count_standar = title_scalar.transform(x_train['title_word_count'].values.reshape(-1, 1))
title_scalar.fit(x_test['title_word_count'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {title_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
test_title_word_count_standar = title_scalar.transform(x_test['title_word_count'].values.reshape(-1, 1))
title_scalar.fit(x_cv['title_word_count'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {title_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
cv_title_word_count_standar = title_scalar.transform(x_cv['quantity'].values.reshape(-1, 1))
print(train_title_word_count_standar.shape, y_train.shape)
print(test_title_word_count_standar.shape, y_test.shape)
print(cv_title_word_count_standar.shape, y_cv.shape)
Standardize Essay_word_count¶
In [111]:
essay_scalar = StandardScaler()
essay_scalar.fit(x_train['essay_word_count'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
train_essay_word_count_standar = essay_scalar.transform(x_train['essay_word_count'].values.reshape(-1, 1))
essay_scalar.fit(x_train['essay_word_count'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
test_essay_word_count_standar = essay_scalar.transform(x_test['essay_word_count'].values.reshape(-1, 1))
essay_scalar.fit(x_cv['essay_word_count'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
cv_essay_word_count_standar = essay_scalar.transform(x_cv['essay_word_count'].values.reshape(-1, 1))
print(train_essay_word_count_standar.shape, y_train.shape)
print(test_essay_word_count_standar.shape, y_test.shape)
print(cv_essay_word_count_standar.shape, y_cv.shape)
Standardize Positive Intensity¶
In [120]:
essay_scalar.fit(x_train['positive'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
train_positive_standar = essay_scalar.transform(x_train['positive'].values.reshape(-1, 1))
essay_scalar.fit(x_train['positive'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
test_positive_standar = essay_scalar.transform(x_test['positive'].values.reshape(-1, 1))
essay_scalar.fit(x_cv['positive'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
cv_positive_standar = essay_scalar.transform(x_cv['positive'].values.reshape(-1, 1))
print(train_positive_standar.shape, y_train.shape)
print(test_positive_standar.shape, y_test.shape)
print(cv_positive_standar.shape, y_cv.shape)
Standarsize Negitive Intensity¶
In [121]:
essay_scalar.fit(x_train['negitive'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
train_negitive_standar = essay_scalar.transform(x_train['negitive'].values.reshape(-1, 1))
essay_scalar.fit(x_train['negitive'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
test_negitive_standar = essay_scalar.transform(x_test['negitive'].values.reshape(-1, 1))
essay_scalar.fit(x_cv['negitive'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
cv_negitive_standar = essay_scalar.transform(x_cv['negitive'].values.reshape(-1, 1))
print(train_negitive_standar.shape, y_train.shape)
print(test_negitive_standar.shape, y_test.shape)
print(cv_negitive_standar.shape, y_cv.shape)
Standardize Neutral Intensity¶
In [122]:
essay_scalar.fit(x_train['neutral'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
train_neutral_standar = essay_scalar.transform(x_train['neutral'].values.reshape(-1, 1))
essay_scalar.fit(x_train['neutral'].values.reshape(-1,1))
test_neutral_standar = essay_scalar.transform(x_test['neutral'].values.reshape(-1, 1))
essay_scalar.fit(x_cv['neutral'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
cv_neutral_standar = essay_scalar.transform(x_cv['neutral'].values.reshape(-1, 1))
print(train_neutral_standar.shape, y_train.shape)
print(test_neutral_standar.shape, y_test.shape)
print(cv_neutral_standar.shape, y_cv.shape)
In [ ]:
Assignment 5: Logistic Regression¶
- [Task-1] Logistic Regression(either SGDClassifier with log loss, or LogisticRegression) on these feature sets
- Set 1: categorical, numerical features + project_title(BOW) + preprocessed_eassay (`BOW with bi-grams` with `min_df=10` and `max_features=5000`)
- Set 2: categorical, numerical features + project_title(TFIDF)+ preprocessed_eassay (`TFIDF with bi-grams` with `min_df=10` and `max_features=5000`)
- Set 3: categorical, numerical features + project_title(AVG W2V)+ preprocessed_eassay (AVG W2V)
- Set 4: categorical, numerical features + project_title(TFIDF W2V)+ preprocessed_essay (TFIDF W2V)
- Hyper paramter tuning (find best hyper parameters corresponding the algorithm that you choose)
- Find the best hyper parameter which will give the maximum AUC value
- Find the best hyper paramter using k-fold cross validation or simple cross validation data
- Use gridsearch cv or randomsearch cv or you can also write your own for loops to do this task of hyperparameter tuning
- Representation of results
- You need to plot the performance of model both on train data and cross validation data for each hyper parameter, like shown in the figure.
- Once after you found the best hyper parameter, you need to train your model with it, and find the AUC on test data and plot the ROC curve on both train and test.
- Along with plotting ROC curve, you need to print the confusion matrix with predicted and original labels of test data points. Please visualize your confusion matrices using seaborn heatmaps.
- You need to plot the performance of model both on train data and cross validation data for each hyper parameter, like shown in the figure.
- [Task-2] Apply Logistic Regression on the below feature set Set 5 by finding the best hyper parameter as suggested in step 2 and step 3.
- Consider these set of features Set 5 :
- school_state : categorical data
- clean_categories : categorical data
- clean_subcategories : categorical data
- project_grade_category :categorical data
- teacher_prefix : categorical data
- quantity : numerical data
- teacher_number_of_previously_posted_projects : numerical data
- price : numerical data
- sentiment score's of each of the essay : numerical data
- number of words in the title : numerical data
- number of words in the combine essays : numerical data
- Conclusion
- You need to summarize the results at the end of the notebook, summarize it in the table format. To print out a table please refer to this prettytable library link
- You need to summarize the results at the end of the notebook, summarize it in the table format. To print out a table please refer to this prettytable library link
Note: Data Leakage
- There will be an issue of data-leakage if you vectorize the entire data and then split it into train/cv/test.
- To avoid the issue of data-leakage, make sure to split your data first and then vectorize it.
- While vectorizing your data, apply the method fit_transform() on you train data, and apply the method transform() on cv/test data.
- For more details please go through this link.
LOGISTIC REGRESSION¶
2.1 Logistic Regression On Set-1¶
Merging all Categorical and Numerical _ SET-1 BOW Encoding¶
In [123]:
from scipy.sparse import hstack
# with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
X_train1 = hstack((train_categories_one_hot,train_sub_categories_one_hot,train_state_one_hot,train_grade_one_hot,
train_teacher_one_hot,train_text_bow, train_titles_bow, train_quantity_standar,
train_prev_proj_standar, train_price_standar, train_title_word_count_standar,
train_essay_word_count_standar, train_positive_standar, train_negitive_standar,
train_neutral_standar)).tocsr()
print(X_train1.shape, y_train.shape)
print(type(X_train1))
In [124]:
X_test1 = hstack((test_categories_one_hot,test_sub_categories_one_hot,test_state_one_hot,test_grade_one_hot,
test_teacher_one_hot,test_text_bow, test_titles_bow, test_quantity_standar,
test_prev_proj_standar, test_price_standar, test_essay_word_count_standar,
test_title_word_count_standar, test_positive_standar, test_negitive_standar,
test_neutral_standar)).tocsr()
print(X_test1.shape, y_test.shape)
print(type(X_test1))
In [125]:
X_cv1 = hstack((cv_categories_one_hot, cv_sub_categories_one_hot, cv_state_one_hot, cv_grade_one_hot,
cv_teacher_one_hot, cv_text_bow, cv_titles_bow, cv_quantity_standar,
cv_prev_proj_standar, cv_price_standar, cv_essay_word_count_standar,
cv_title_word_count_standar, cv_positive_standar, cv_negitive_standar,
cv_neutral_standar)).tocsr()
print(X_cv1.shape, y_cv.shape)
print(type(X_cv1))
In [126]:
print(X_train1.shape, y_train.shape)
Hyperparameter Tunning¶
In [127]:
from sklearn.model_selection import train_test_split
from sklearn.grid_search import GridSearchCV
from sklearn.model_selection import GridSearchCV
#from sklearn.datasets import *
from sklearn.linear_model import LogisticRegression
In [128]:
import math
log_parameters = []
parameters = [10**4, 10**3, 10**2, 10**1, 10**0, 10**-1, 10**-2, 10**-3, 10**-4]
for i in parameters:
j = math.log(i)
log_parameters.append(j)
print(log_parameters)
In [129]:
parameters = {'C':[10**4, 10**3, 10**2, 10**1, 10**0, 10**-1, 10**-2, 10**-3, 10**-4]}
classifier = GridSearchCV(LogisticRegression(), parameters, cv= 10, scoring='roc_auc')
classifier.fit(X_train1, y_train)
train_auc= classifier.cv_results_['mean_train_score']
train_auc_std= classifier.cv_results_['std_train_score']
cv_auc = classifier.cv_results_['mean_test_score']
cv_auc_std= classifier.cv_results_['std_test_score']
In [130]:
#classifier = GridSearchCV(LogisticRegression(), parameters, cv= 10, scoring='roc_auc')
#classifier.fit(X_train1, y_train)
#train_auc= classifier.cv_results_['mean_train_score']
#train_auc_std= classifier.cv_results_['std_train_score']
#cv_auc = classifier.cv_results_['mean_test_score']
#cv_auc_std= classifier.cv_results_['std_test_score']
#how to draw roc curve for logistic regression gridsearchcv
#https://stackoverflow.com/questions/48796282/how-to-visualize-dependence-of-model-performance-alpha-with-matplotlib/48803361#48803361
#https://stackoverflow.com/questions/42894871/how-to-plot-multiple-roc-curves-in-one-plot-with-legend-and-auc-scores-in-python
#
plt.plot(log_parameters, train_auc, label='Train AUC')
plt.plot(log_parameters, cv_auc, label='CV AUC')
plt.scatter(log_parameters, train_auc, label='Train AUC points')
plt.scatter(log_parameters, cv_auc, label='CV AUC points')
plt.legend()
plt.xlabel("log_C : Hyperparameter")
plt.ylabel("ACCURACY")
plt.title("C: Hyperparameter v/s AUC")
plt.grid()
plt.show()
In [131]:
def batch_predict(clf, data):
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
y_data_pred = []
tr_loop = data.shape[0] - data.shape[0]%1000
# consider you X_tr shape is 49041, then your cr_loop will be 49041 - 49041%1000 = 49000
# in this for loop we will iterate unti the last 1000 multiplier
for i in range(0, tr_loop, 1000):
y_data_pred.extend(clf.predict_proba(data[i:i+1000])[:,1])
# we will be predicting for the last data points
y_data_pred.extend(clf.predict_proba(data[tr_loop:])[:,1])
return y_data_pred
In [132]:
# The Hyperpaameter cannot be -VE since we taking the least values as Hyperparameter
best_c_1 = 0.01
In [133]:
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc
LGR_bow = LogisticRegression(C = best_c_1)
LGR_bow.fit(X_train1, y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
y_train_pred = batch_predict(LGR_bow, X_train1)
y_test_pred = batch_predict(LGR_bow, X_test1)
train_fpr, train_tpr, tr_thresholds = roc_curve(y_train, y_train_pred)
test_fpr, test_tpr, te_thresholds = roc_curve(y_test, y_test_pred)
plt.plot(train_fpr, train_tpr, label="Train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="Test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.xlabel("True Positive Rate(TPR)")
plt.ylabel("False Positive Rate(FPR)")
plt.title("AUC")
plt.grid()
plt.show()
Confusion Matrix¶
In [134]:
def predict(proba, threshould, fpr, tpr):
t = threshould[np.argmax(fpr*(1-tpr))]
print("the maximum value of tpr*(1-fpr)", max(tpr*(1-fpr)), "for threshold", np.round(t,3))
predictions = []
for i in proba:
if i>=t:
predictions.append(1)
else:
predictions.append(0)
return predictions
Train Data¶
In [135]:
from sklearn.metrics import confusion_matrix
import seaborn as sea
print("Train confusion matrix")
print(confusion_matrix(y_train, predict(y_train_pred, tr_thresholds, train_fpr, train_fpr)))
In [136]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[136]:
Test Data¶
In [137]:
print("Test confusion matrix")
print(confusion_matrix(y_test, predict(y_test_pred, tr_thresholds, test_fpr, test_fpr)))
In [138]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[138]:
2.2 Merging all Categorical and Numerical _ SET-2 TF-IDF Encoding¶
In [148]:
X_train2 = hstack((train_categories_one_hot, train_sub_categories_one_hot, train_state_one_hot, train_grade_one_hot,
train_teacher_one_hot,train_text_tfidf, train_title_tfidf, train_quantity_standar,
train_prev_proj_standar, train_price_standar, train_title_word_count_standar,
train_essay_word_count_standar, train_positive_standar, train_negitive_standar,
train_neutral_standar)).tocsr()
print(X_train2.shape, y_train.shape)
print(type(X_train2))
In [152]:
X_test2 = hstack((test_categories_one_hot,test_sub_categories_one_hot,test_state_one_hot,test_grade_one_hot,
test_teacher_one_hot,test_text_tfidf, test_title_tfidf, test_quantity_standar,
test_prev_proj_standar, test_price_standar,test_title_word_count_standar,
test_essay_word_count_standar, test_positive_standar, test_negitive_standar,
test_neutral_standar)).tocsr()
print(X_test2.shape, y_test.shape)
print(type(X_test2))
In [153]:
X_cv2 = hstack((cv_categories_one_hot,cv_sub_categories_one_hot,cv_state_one_hot,cv_grade_one_hot,
cv_teacher_one_hot,cv_text_tfidf, cv_title_tfidf, cv_quantity_standar,
cv_prev_proj_standar, cv_price_standar, cv_title_word_count_standar,
cv_essay_word_count_standar, cv_positive_standar, cv_negitive_standar,
cv_neutral_standar)).tocsr()
print(X_cv2.shape, y_cv.shape)
print(type(X_cv2))
Hyperparameter Tunning¶
In [143]:
#parameters = {'C':[0.5, 0.25, 0.125, 0.0625, 0.03125, 0.015625, 0.0078125, 0.0039, 0.00195, 0.0009 ]}
parameters = {'C':[10**4, 10**3, 10**2, 10**1, 10**0, 10**-1, 10**-2, 10**-3, 10**-4]}
classifier = GridSearchCV(LogisticRegression(), parameters, cv= 10, scoring='roc_auc')
classifier.fit(X_train2, y_train)
train_auc= classifier.cv_results_['mean_train_score']
train_auc_std= classifier.cv_results_['std_train_score']
cv_auc = classifier.cv_results_['mean_test_score']
cv_auc_std= classifier.cv_results_['std_test_score']
In [144]:
plt.plot(log_parameters, train_auc, label='Train AUC')
plt.plot(log_parameters, cv_auc, label='CV AUC')
plt.scatter(log_parameters, train_auc, label='Train AUC points')
plt.scatter(log_parameters, cv_auc, label='CV AUC points')
plt.legend()
plt.xlabel("log_C : Hyperparameter")
plt.ylabel("ACCURACY")
plt.title("C: Hyperparameter v/s AUC")
plt.grid()
plt.show()
In [146]:
best_c_2 = 0.1
In [154]:
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc
LGR_tfidf = LogisticRegression(C = best_c_2)
LGR_tfidf.fit(X_train2, y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
y_train_pred = batch_predict(LGR_bow, X_train2)
y_test_pred = batch_predict(LGR_bow, X_test2)
train_fpr, train_tpr, tr_thresholds = roc_curve(y_train, y_train_pred)
test_fpr, test_tpr, te_thresholds = roc_curve(y_test, y_test_pred)
plt.plot(train_fpr, train_tpr, label="Train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="Test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.xlabel("True Positive Rate(TPR)")
plt.ylabel("False Positive Rate(FPR)")
plt.title("AUC")
plt.grid()
plt.show()
Train Confusing Matrix¶
In [155]:
from sklearn.metrics import confusion_matrix
import seaborn as sea
print("Train confusion matrix")
print(confusion_matrix(y_train, predict(y_train_pred, tr_thresholds, train_fpr, train_fpr)))
In [156]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[156]:
Test Confusion Matrix¶
In [157]:
print("Test confusion matrix")
print(confusion_matrix(y_test, predict(y_test_pred, tr_thresholds, test_fpr, test_fpr)))
In [158]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[158]:
In [ ]:
2.3 Merging all Categorical and Numerical _ SET-3 AVG-W2V Encoding¶
In [159]:
from scipy.sparse import hstack
# with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
X_train3 = hstack((train_categories_one_hot,train_sub_categories_one_hot,train_state_one_hot,train_grade_one_hot,
train_teacher_one_hot, train_title_avg_w2v_vectors, train_avg_w2v_vectors, train_quantity_standar,
train_prev_proj_standar, train_price_standar, train_positive_standar, train_negitive_standar,
train_neutral_standar, train_title_word_count_standar, train_essay_word_count_standar)).tocsr()
print(X_train3.shape, y_train.shape)
print(type(X_train3))
In [162]:
X_test3 = hstack((test_categories_one_hot,test_sub_categories_one_hot,test_state_one_hot,test_grade_one_hot,
test_teacher_one_hot, test_title_avg_w2v_vectors, test_avg_w2v_vectors, test_quantity_standar,
test_prev_proj_standar, test_price_standar, test_positive_standar, test_negitive_standar,
test_neutral_standar, test_title_word_count_standar, test_essay_word_count_standar)).tocsr()
print(X_test3.shape, y_test.shape)
print(type(X_test3))
In [164]:
X_cv3 = hstack((cv_categories_one_hot, cv_sub_categories_one_hot, cv_state_one_hot, cv_grade_one_hot,
cv_teacher_one_hot, cv_title_avg_w2v_vectors, cv_avg_w2v_vectors, cv_quantity_standar,
cv_prev_proj_standar, cv_price_standar, cv_positive_standar, cv_negitive_standar,
cv_neutral_standar, cv_title_word_count_standar, cv_essay_word_count_standar)).tocsr()
print(X_cv3.shape, y_cv.shape)
print(type(X_cv3))
In [165]:
print(X_train3.shape, y_train.shape)
Hyperparameter Tunning¶
In [166]:
parameters = {'C':[10**4, 10**3, 10**2, 10**1, 10**0, 10**-1, 10**-2, 10**-3, 10**-4]}
classifier = GridSearchCV(LogisticRegression(), parameters, cv= 10, scoring='roc_auc')
classifier.fit(X_train3, y_train)
train_auc= classifier.cv_results_['mean_train_score']
train_auc_std= classifier.cv_results_['std_train_score']
cv_auc = classifier.cv_results_['mean_test_score']
cv_auc_std= classifier.cv_results_['std_test_score']
In [167]:
plt.plot(log_parameters, train_auc, label='Train AUC')
plt.plot(log_parameters, cv_auc, label='CV AUC')
plt.scatter(log_parameters, train_auc, label='Train AUC points')
plt.scatter(log_parameters, cv_auc, label='CV AUC points')
plt.legend()
plt.xlabel("log_C : Hyperparameter")
plt.ylabel("ACCURACY")
plt.title("C: Hyperparameter v/s AUC")
plt.grid()
plt.show()
In [168]:
best_c_3 = 1
In [169]:
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve
from sklearn.metrics import roc_curve, auc
LGR_avgw2v = LogisticRegression(C = best_c_3)
LGR_avgw2v.fit(X_train3, y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
y_train_pred = batch_predict(LGR_avgw2v, X_train3)
y_test_pred = batch_predict(LGR_avgw2v, X_test3)
train_fpr, train_tpr, tr_thresholds = roc_curve(y_train, y_train_pred)
test_fpr, test_tpr, te_thresholds = roc_curve(y_test, y_test_pred)
plt.plot(train_fpr, train_tpr, label="Train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="Test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.xlabel("True Positive Rate(TPR)")
plt.ylabel("False Positive Rate(FPR)")
plt.title("AUC")
plt.grid()
plt.show()
Train Confusion Matrix¶
In [170]:
from sklearn.metrics import confusion_matrix
import seaborn as sea
print("Train confusion matrix")
print(confusion_matrix(y_train, predict(y_train_pred, tr_thresholds, train_fpr, train_fpr)))
In [171]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[171]:
Test Confusion Matrix¶
In [172]:
print("Test confusion matrix")
print(confusion_matrix(y_test, predict(y_test_pred, tr_thresholds, test_fpr, test_fpr)))
In [173]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[173]:
2.4 Merging all Categorical and Numerical _ SET-4 TFIDF-W2V Encoding¶
In [174]:
from scipy.sparse import hstack
# with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
X_train4 = hstack((train_categories_one_hot,train_sub_categories_one_hot,train_state_one_hot,train_grade_one_hot,
train_teacher_one_hot, train_title_tfidf_w2v_vectors, train_essay_tfidf_w2v_vectors,
train_quantity_standar, train_prev_proj_standar, train_price_standar,train_positive_standar,
train_negitive_standar, train_neutral_standar, train_title_word_count_standar,
train_essay_word_count_standar)).tocsr()
print(X_train4.shape, y_train.shape)
print(type(X_train4)) #train_title_tfidf_w2v_vectors train_essay_tfidf_w2v_vectors
In [187]:
X_test4 = hstack((test_categories_one_hot,test_sub_categories_one_hot,test_state_one_hot,test_grade_one_hot,
test_teacher_one_hot, test_title_tfidf_w2v_vectors, test_essay_tfidf_w2v_vectors,
test_quantity_standar, test_prev_proj_standar, test_price_standar, test_positive_standar,
test_negitive_standar, test_neutral_standar, test_title_word_count_standar,
test_essay_word_count_standar)).tocsr()
print(X_test4.shape, y_test.shape)
print(type(X_test4)) #train_title_tfidf_w2v_vectors train_essay_tfidf_w2v_vectors
In [188]:
X_cv4 = hstack((cv_categories_one_hot, cv_sub_categories_one_hot, cv_state_one_hot, cv_grade_one_hot,
cv_teacher_one_hot, cv_title_tfidf_w2v_vectors, cv_essay_tfidf_w2v_vectors,
cv_quantity_standar, cv_prev_proj_standar, cv_price_standar,cv_positive_standar,
cv_negitive_standar, cv_neutral_standar, cv_title_word_count_standar,
cv_essay_word_count_standar)).tocsr()
print(X_cv4.shape, y_cv.shape)
print(type(X_cv4)) #train_title_tfidf_w2v_vectors train_essay_tfidf_w2v_vectors
In [190]:
print(X_train4.shape, y_train.shape)
Hyperparameter Tunning¶
In [191]:
parameters = {'C':[10**4, 10**3, 10**2, 10**1, 10**0, 10**-1, 10**-2, 10**-3, 10**-4]}
classifier = GridSearchCV(LogisticRegression(), parameters, cv= 10, scoring='roc_auc')
classifier.fit(X_train4, y_train)
train_auc= classifier.cv_results_['mean_train_score']
train_auc_std= classifier.cv_results_['std_train_score']
cv_auc = classifier.cv_results_['mean_test_score']
cv_auc_std= classifier.cv_results_['std_test_score']
In [192]:
plt.plot(log_parameters, train_auc, label = "Train_AUC")
plt.plot(log_parameters, cv_auc, label = "CV_AUC")
plt.scatter(log_parameters, train_auc, label = "Train AUC Points")
plt.scatter(log_parameters, cv_auc, label = "CV AUC Points")
plt.xlabel("C: Hyperparameter")
plt.ylabel("Accuracy")
plt.title("C: Hyperparameter v/s Accuracy")
plt.legend()
plt.grid()
plt.show()
In [193]:
best_c_4 = 0.2
In [194]:
from sklearn.metrics import roc_curve, auc
LGR_tfidfw2v = LogisticRegression(C = best_c_4)
LGR_tfidfw2v.fit(X_train4, y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
y_train_pred = batch_predict(LGR_tfidfw2v, X_train4)
y_test_pred = batch_predict(LGR_tfidfw2v, X_test4)
train_fpr, train_tpr, tr_thresholds = roc_curve(y_train, y_train_pred)
test_fpr, test_tpr, te_thresholds = roc_curve(y_test, y_test_pred)
plt.plot(train_fpr, train_tpr, label="Train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="Test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.xlabel("True Positive Rate(TPR)")
plt.ylabel("False Positive Rate(FPR)")
plt.title("AUC")
plt.grid()
plt.show()
Train Confusion Matrix¶
In [195]:
from sklearn.metrics import confusion_matrix
import seaborn as sea
print("Train confusion matrix")
print(confusion_matrix(y_train, predict(y_train_pred, tr_thresholds, train_fpr, train_fpr)))
In [196]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[196]:
Test Confusion Matrix¶
In [197]:
print("Test confusion matrix")
print(confusion_matrix(y_test, predict(y_test_pred, tr_thresholds, test_fpr, test_fpr)))
In [198]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[198]:
2.5 Apply Logistic Regression on Set-5¶
[Task-2] Apply Logistic Regression on the below feature set Set 5 by finding the best hyper parameter GridSearch </br>
Consider these set of features for Set 5 in Assignment:
categorical dataschool_state clean_categories....clean_subcategories....project_grade_category....teacher_prefix
numerical data quantity....teacher_number_of_previously_posted_projects....price
New Features
sentiment score's of each of the essay : numerical data
number of words in the title : numerical data
number of words in the combine essays : numerical data
In [182]:
X_train5 = hstack((train_categories_one_hot, train_sub_categories_one_hot, train_grade_one_hot,
train_teacher_one_hot, train_quantity_standar, train_prev_proj_standar, train_price_standar,
train_title_word_count_standar, train_essay_word_count_standar, train_positive_standar,
train_negitive_standar, train_neutral_standar)).tocsr()
print(X_train5.shape, y_train.shape)
print(type(X_train5))
In [184]:
X_test5 = hstack((test_categories_one_hot, test_sub_categories_one_hot, test_grade_one_hot,
test_teacher_one_hot, test_quantity_standar, test_prev_proj_standar, test_price_standar,
test_title_word_count_standar, test_essay_word_count_standar, test_positive_standar,
test_negitive_standar, test_neutral_standar)).tocsr()
print(X_test5.shape, y_train.shape)
print(type(X_test5))
In [185]:
X_cv5 = hstack((cv_categories_one_hot, cv_sub_categories_one_hot, cv_grade_one_hot,
cv_teacher_one_hot, cv_quantity_standar, cv_prev_proj_standar, cv_price_standar,
cv_title_word_count_standar, cv_essay_word_count_standar, cv_positive_standar,
cv_negitive_standar, cv_neutral_standar)).tocsr()
print(X_cv5.shape, y_train.shape)
print(type(X_cv5))
In [199]:
y_trainn = y_train[0:24155,]
print(y_trainn.shape)
Hyperparameter Tunning¶
In [202]:
parameters = {'C':[10**4, 10**3, 10**2, 10**1, 10**0, 10**-1, 10**-2, 10**-3, 10**-4]}
classifier = GridSearchCV(LogisticRegression(), parameters, cv= 10, scoring='roc_auc')
classifier.fit(X_train5, y_train)
train_auc= classifier.cv_results_['mean_train_score']
train_auc_std= classifier.cv_results_['std_train_score']
cv_auc = classifier.cv_results_['mean_test_score']
cv_auc_std= classifier.cv_results_['std_test_score']
In [203]:
plt.plot(log_parameters, train_auc, label = "Train_AUC")
plt.plot(log_parameters, cv_auc, label = "CV_AUC")
plt.scatter(log_parameters, train_auc, label = "Train AUC Points")
plt.scatter(log_parameters, cv_auc, label = "CV AUC Points")
plt.xlabel("C: Hyperparameter")
plt.ylabel("Accuracy")
plt.title("C: Hyperparameter v/s Accuracy")
plt.legend()
plt.grid()
plt.show()
In [204]:
best_c_5 = 0.5
In [205]:
from sklearn.metrics import roc_curve, auc
LGR_tfidfw2v = LogisticRegression(C = best_c_5)
LGR_tfidfw2v.fit(X_train5, y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
y_train_pred = batch_predict(LGR_tfidfw2v, X_train5)
y_test_pred = batch_predict(LGR_tfidfw2v, X_test5)
train_fpr, train_tpr, tr_thresholds = roc_curve(y_train, y_train_pred)
test_fpr, test_tpr, te_thresholds = roc_curve(y_test, y_test_pred)
plt.plot(train_fpr, train_tpr, label="Train AUC ="+str(auc(train_fpr, train_tpr)))
plt.plot(test_fpr, test_tpr, label="Test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.xlabel("True Positive Rate(TPR)")
plt.ylabel("False Positive Rate(FPR)")
plt.title("AUC")
plt.grid()
plt.show()
Train Confusion Matrix¶
In [206]:
from sklearn.metrics import confusion_matrix
import seaborn as sea
print("Train confusion matrix")
print(confusion_matrix(y_train, predict(y_train_pred, tr_thresholds, train_fpr, train_fpr)))
In [207]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[207]:
Test Confusion Matrix¶
In [208]:
print("Test confusion matrix")
print(confusion_matrix(y_test, predict(y_test_pred, tr_thresholds, test_fpr, test_fpr)))
In [209]:
train_confusion_matrix = pd.DataFrame(confusion_matrix(y_test,predict(y_test_pred, tr_thresholds,
test_fpr,test_fpr)), range(2),range(2))
sea.set(font_scale=1.4)
sea.heatmap(train_confusion_matrix, annot = True, annot_kws={"size":16}, fmt = 'd')
plt.xlabel("Predicted Value")
plt.ylabel("True Value")
plt.title("Test Confusion Matix")
Out[209]:
3. Conclusion
In [210]:
# Please compare all your models using Prettytable library
# http://zetcode.com/python/prettytable/
from prettytable import PrettyTable
TB = PrettyTable()
TB.field_names = ["Vectorizer", "C:Hyperparameter", "Train_AUC", "Test_Auc"]
TB.title = "Logistic Regression"
TB.add_row(["BOW-Model", 0.01, 0.78,0.68])
TB.add_row(["TFIDF-Model", 0.1, 0.70, 0.66])
TB.add_row(["AvgW2v-Model", 1.0, 0.74, 0.72])
TB.add_row(["Tf-Idf-Model", 0.2, 0.73, 0.71])
TB.add_row(["NUM_Features-Model", 0.5, 0.63, 0.61])
print(TB)
Observations:¶
From above we can the BOW and TFIDF encodding contain the Project_Essays and Project_titles in those models. So, that we can say that Text Data also plays major role in predicting the output. The Set 5 which we built model on numerical features only performs badly compared to all the Models which having text data.